Summary Eating disorders (EDs) typically onset in adolescence at a time of gonadal hormone changes and rapid brain development. EDs characterized by extreme dietary restriction and/or excessive exercise (ED-R/E) and high drive for thinness are associated with cognitive inflexibility (Cognitive Flexibility), reduced responsiveness to reward (Initial Response to Reward), and altered reward valuation (Delay), which contribute to maintenance of illness and poor outcomes. Hypoestrogenemia is common in ED-R/E (~60%), and in other conditions has been linked to cognitive inflexibility and altered reward responsiveness and valuation. Clarifying the link between estrogen status, Cognitive Flexibility, Initial Response to Reward and Delay, and ED pathology may facilitate identification of novel treatment targets to improve outcomes via an experimental therapeutics approach. Our preliminary data indicate: (i) abnormalities in RDoC domains of Cognitive and Positive Valence systems in hypoestrogenic adolescents/ young adults (independent of weight) compared to normo-estrogenic controls, and (ii) that hypoestrogenemia is associated with reduced Cognitive Flexibility and Initial Response to Reward (neural response to palatable food images), altered Delay (increased preference for larger delayed over immediate smaller rewards), and increased ED pathology. Estrogen deficiency may thus play a key mechanistic role in maintenance of ED-R/E by acting on these RDoC domains. Importantly, hypogonadal adolescents/young women are commonly treated with estrogen replacement for other (e.g. bone) outcomes, and data from our team and others demonstrate that estrogen replacement also improves Cognitive Flexibility, Initial Response to Reward and Delay. Further, our data show that (i) long-term estrogen replacement improves ED pathology and food intake, and (ii) improved Cognitive Flexibility following estrogen replacement predicts improved ED pathology. Published work in other hypogonadal states shows that even short-term (8-12 weeks) estrogen/estrogen agonist administration can alter cognitive flexibility and reward processing. It is now critical to examine whether estrogen deficiency contributes to dysfunction across Cognitive and Positive Valence RDoC domains in ED-R/E, and whether correcting estrogen deficiency improves ED pathology via its impact on these domains. To fill this gap, we propose using physiologic estrogen replacement as a mechanistic probe in ED-R/E. We will randomize 120 hypoestrogenemic females with ED-R/E (ages 16-26) to a 12-week challenge of physiologic estrogen or placebo to evaluate: effects on RDoC subconstructs (Updating, Representation and Maintenance i.e. Cognitive Flexibility; Initial Response to Reward; and Delay) at 8 weeks; ED pathology at 12 weeks; and determine whether 8-week changes in RDoC subconstructs mediate the 12-week improvement in ED pathology. We hypothesize that in ED-R/E, correcting estrogen deficiency will improve Cognitive Flexibility, Initial Response to Reward and Delay, and ED pathology; and that improvement in ED pathology will be mediated by changes in these RDoC subconstructs.